Internet Scofflaw

StrategyPage reports:

A recent analysis of web sites pushing malware (software that helps hackers steal data) revealed that half of them are connected with just ten ISPs (Internet Service Providers), and six of those ISPs are in China. This came as no surprise, as China has become the favorite hideout for Internet criminals.

There’s just one catch. The Chinese Internet is highly policed by a special force of 30,000 secret police technicians. On the Chinese Internet, you don’t do something the government does not want, at least not for long. So how do these criminals manage to survive on such a heavily policed portion of the Internet? It’s no secret that a lot of Internet mischief comes out of China, with the tacit approval of the Chinese government.

The story doesn’t link to the study, but my web search found this post.

(Via Instapundit.)

Google suspends several anti-Obama blogs, due to unsupported allegations of spam. (Via Instapundit.)

This is a good example of why I’m so troubled by Google’s near-monopoly over access to information on the Internet. If Google decides to distort its results, for some political or business purpose, how will people know? This is no abstract worry, either, as Google has already done so, and not only in China. (For example, Google News includes highly dubious “news” sources of the leftist persuasion, but you’d be hard-pressed to find an example from the right.)

In this case, Google is following bad procedures — in the very least — if obviously non-spam blogs are being suspended (and worse, not being reinstated in a timely fashion). But it’s more troublesome than that, because Google is deliberately opaque about what its procedures are. Consequently, if they bias their procedures — or just violate them — for some political or business purpose, we have no way of knowing.

Probably Google is simply utilizing a bad algorithm here that has been exploited by pro-Obama vandals, but given Google’s opaqueness, we can judge the incident only by its outcome. And the outcomes of Google incidents, when they have political implications, usually seem to be in line with the company’s public-record political preferences. (I only say “usually” as a hedge; I know of no counterexamples.)

Obviously, a big part of Google’s procedures is their page rank algorithm, which is an important trade secret. But if they can’t be transparent in that area, they need to make a special effort in other areas. It also wouldn’t hurt for them to try to build bridges with people outside the political left.

UPDATE: The NYT picks up the story.

When doing a news search for the previous post on “gaza truce fails“, two of the top five “news” articles were: The Nation, Pakistani edition (because the American version isn’t extreme enough, I guess), and Socialist Worker Online.

These are “news” sources?!  I would like to hear someone from Google try to explain this.

Next Big Future has an update on the uranium hydride reactor, which Hyperion, a startup company, is preparing to bring to market in 2012.  The reactor would be installed right where the power is needed, and would be fully contained with no serviceable parts, making it more akin to a battery than a reactor.

Hyperion advertises that its reactor should last 7-10 years and would cost $1400 per kilowatt during that time.  Taking the conservative 7-year duration, that works out to 2.28 cents per kilowatt-hour (or a little more if you take into account present value).  By comparison, I’m paying 8.41 cents per kilowatt-hour in Pittsburgh right now.

Their first installations will be at oil sand and oil shale facilities, where they hope their safety record will make people comfortable with installing them closer to home.

ASIDE: My understanding is that the hydrogen atoms in uranium hydride damp the neutrons, thereby slowing the fission reaction.  That’s bad for weapons but good for energy generation.

This sounds cool:

Defense contractor Northrop Grumman won a $6.7 million contract to develop brain-wave binoculars.

The Defense Advanced Research Projects Agency, better known as DARPA, awarded the contract to develop intelligent binoculars that would help soldiers detect threats from miles away. The defense contractor says electrodes placed on the scalp will record the user’s electrical brain activity. Responses will train the system over time to recognize actual threats at greater distances than conventional binoculars.

The system would use a custom helmet equipped with wide-angle binoculars capable of producing high-resolution images and electroencephalogram, or EEG, electrodes. Researchers hope to tap into the brain’s ability to spot patterns and movement.

The article doesn’t say how it works, and seems to imply that the device would tap into subconscious brain activity.  That would definitely be cool, but it seems more likely to me that the brain-wave monitoring is used to identify threats that the wearer is aware of, and then a machine-learning algorithm uses that information to learn to recognize threats from video input.

The Economist reports that police are looking for ways to catch fugitives in cars without dangerous high-speed chases:

One way to avoid the need for chases would be to track felonious vehicles electronically, instead of running after them. StarChase, a company based in Virginia Beach, Virginia, has developed a way to do just that. A pneumatic cannon is mounted on the pursuit car. With the help of a guiding laser, it shoots a satellite-based tracking device, smothered in epoxy goo, onto the target vehicle, allowing the police to track the suspect without endangering the public. The Los Angeles Police Department plans to deploy the system this year.

Sounds good, and they mention several other interesting ideas as well. This one, however, I’m not wild about:

From the police’s point of view, however, it would be better if they could actually stop a runaway car by satellite, not just track it. General Motors plans to allow them to do just that. From September its OnStar service, which provides navigation and emergency services to drivers, will include a system called Stolen Vehicle Slowdown. Police who believe a car to be stolen can ask an OnStar operator to disable its accelerator, while leaving the steering and brakes in working order. Some people worry that hackers might take over the system. But Chet Huber, OnStar’s boss, reckons that the benefits outweigh the risks.

Yes, the hackers are a worry, but a greater worry is the government. Once the government starts controlling our vehicles, how long will they limit themselves to this narrow purpose? Not long, history tells us.

ASIDE: The Long Run, one of my favorite novels, paints a picture of an oppressive world government that uses control over vehicles as one of its tools for controlling the populace.

Two months ago, I linked to a chronology of missile defense tests at the Missile Defense Advocacy Alliance web site. Unfortunately, that link went stale, so I decided to assemble a chronology of my own by correlating dates from the MDAA with press releases at the Missile Defense Agency. Since the first missile defense system was ordered deployed in December 2002, there have been 23 tests reported by the MDA:

• Jun 5, 2008: Aegis/SM-2 test successful (pdf).
• Dec 17 2007: Japanese Aegis/SM-3 test successful (pdf).
• Dec 4, 2007: NCADE test successful (pdf).
• Nov 6, 2007: Aegis/SM-3 test successful (pdf).
• Oct 27, 2007: THAAD test successful (pdf).
• Sep 28, 2007: GBMD test successful (pdf).
• Jun 22, 2007: Aegis/SM-3 test successful (pdf).
• Apr 26, 2007: Aegis/SM-3 test successful (pdf).
• Apr 6, 2007: THAAD test successful (pdf).
• Jan 27, 2007: THAAD test successful (pdf).
• Dec 7, 2006: Aegis/SM-3 test unsuccessful (pdf).
• Sep 1, 2006: GBMD test successful (pdf).
• Jul 12, 2006: THAAD test successful (pdf).
• Jun 22, 2006: Aegis/SM-3 test successful (pdf).
• May 24, 2006: Aegis/SM-2 test successful (pdf).
• Nov 17, 2005: Aegis/SM-3 test successful (pdf).
• Sep 8, 2005: PAC-3 test successful (pdf).
• Feb 24, 2005: Aegis/SM-3 test successful (pdf).
• Nov 18 2004: PAC-3 test successful (pdf).
• Sep 2, 2004: PAC-3 test successful (pdf).
• Mar 4, 2004: PAC-3 test successful.
• Dec 11, 2003: Aegis/SM-3 test successful.
• Jun 18, 2003: Aegis/SM-3 test unsuccessful.

For those keeping score, that’s 23 successes and 2 failures. Both failures were for the Aegis/SM-3; one in 2006 when an interceptor failed to fire, the other back in 2003 when an interceptor missed its target.

GMBD (pdf) (ground-based midcourse defense) is the best-known, long-range system, designed to destroy missiles at a distance.  THAAD (pdf) (terminal high-altitude area defense) is a portable ground-based system.  Aegis/SM-2 and SM-3 (pdf) are sea-based systems (using two different missiles).  NCADE (pdf) (net-centric air defense element) is an air-based system, designed to destroy missiles during their boost phase.  PAC-3 (pdf) (Patriot advanced capability) is an evolution of the Patriot system, designed for short-range interceptions.

ASIDE: The MDAA says there have been four more PAC-3 tests (all successful) than the MDA has reported. (The MDA stopped issuing press releases related to the PAC-3 system after its successful test in September 2005.) If we count those, the record goes to 27-2.

Potentially exciting news:

“Ten years ago I could never have imagined I’d be doing this,” says Greg Pal, 33, a former software executive, as he squints into the late afternoon Californian sun. “I mean, this is essentially agriculture, right? But the people I talk to – especially the ones coming out of business school – this is the one hot area everyone wants to get into.”

He means bugs. To be more precise: the genetic alteration of bugs – very, very small ones – so that when they feed on agricultural waste such as woodchips or wheat straw, they do something extraordinary. They excrete crude oil.

Unbelievably, this is not science fiction. Mr Pal holds up a small beaker of bug excretion that could, theoretically, be poured into the tank of the giant Lexus SUV next to us. Not that Mr Pal is willing to risk it just yet. He gives it a month before the first vehicle is filled up on what he calls “renewable petroleum”. After that, he grins, “it’s a brave new world”. . .

What is most remarkable about what they are doing is that instead of trying to reengineer the global economy – as is required, for example, for the use of hydrogen fuel – they are trying to make a product that is interchangeable with oil. The company claims that this “Oil 2.0” will not only be renewable but also carbon negative – meaning that the carbon it emits will be less than that sucked from the atmosphere by the raw materials from which it is made.

(Via Instapundit.)

This would be terrific news if it worked out.  Of course, this isn’t the only way to make artificial petroleum; for over eighty years it’s been possible (via the Fischer-Tropsch process) to make oil from coal, natural gas, or (in principle) biomass.  (South Africa manufactures most of its diesel fuel that way.)  The question is whether the process is cost-effective enough to compete with conventional oil wells.

Still, this work sounds like it has two advantages over the Fischer-Tropsch process.  It generates petroleum, which could be used in our existing infrastructure.  And, there’s the environmental advantage of making fuel from agricultural waste rather than coal or natural gas.

iPhone 3G users will have to use AT&T’s 3G network, which is not as extensive as some. In fact, AT&T won’t give you a nationwide map; you need to select particular areas to see what 3G coverage they have. But iPhone Atlas features a nationwide map that someone cobbled together from AT&T’s little maps.  (Via End User, via Instapundit.)

One detail left out in yesterday’s iPhone 3G extravaganza was the cost of a service plan.  Since the new iPhone moves up to 3G from EDGE, one could expect a hike in price.  Now Fox News reports that the extra cost will be $10 per month.

That’s less than I expected, but, oddly, the article makes it out to be bad news.

An engrossing true story, at Wired.  (Via LGF.)

The Department of Homeland Security is experimenting with the laser system, called Project Chloe (after the “24″ character).   The system would sense a missile launch and fire its laser to jam the missile’s heat-seeking guidance system.  Flying at 60,000 feet, the DHS hopes that a single such system could protect all commercial airports in LA County.

Fox News reports on a terrific new water purification tool:

On the outside, it looks like an ordinary sports bottle. On the inside, there’s a miracle: an extremely advanced filtration system that makes murky water filled with deadly viruses and bacteria completely clean in just seconds.

The Lifesaver removes 99.999 percent of water-borne pathogens and reduces heavy metals like lead, meaning even the filthiest water can be cleaned — immediately.

It will be a boon to soldiers in the field, so it’s winning accolades from the military.

It also stands to revolutionize humanitarian aid. It could be the first weapon in the fight against disease after a natural disaster, like the one in Myanmar this week. . .

Outdoor enthusiasts may find it useful, but the Lifesaver is perfect for the military. The bottle is designed to “scoop and go,” so soldiers won’t have to carry the added weight of clean bottled water. They can pick some up out of any source and keep moving.

As an added bonus, the bottle can shoot a pressurized jet of water from any angle, which will be useful for washing wounds free of contaminants and debris.

There’s video at the link.

The debate over net neutrality is frustrating, since so few people seem to know what they are talking about. This AP article (”FCC chief says no need for new regulation of the Internet”) is a good example:

The hearing was called at a time when the issue of “network neutrality”—the principle that people should be able to go where they choose on the Internet without interference from network owners—has heated up.

(Via Instapundit.)

This is exactly not what network neutrality is about. The main thing to remember is the Internet is not, contrary to popular opinion, a bunch of wires. The Internet is a protocol. Specifically, the Internet Protocol (IP) is a way to route packets over a variety of networks.

An important property of IP is it provides best effort delivery. That means that sometimes it drops packets, typically when it gets too much traffic. Indeed, there is no way to prevent this in a packet-switched network, unless you can prevent routers from getting too much traffic, which IP does not do. If you want reliable delivery, you need to layer another protocol (such as TCP) on top of IP.

IP does not dictate any rules regarding how a router chooses which packets to drop. (This is what network neutrality supporters want to change.) Typically it chooses them arbitrarily. But, it could do something more sensible, based on the nature of the packets. Some packets are more important than others. For example, a video stream typically contains some keyframes and various other frames that depend on the latest keyframe. Dropping one of the latter frames is no big deal, but dropping a keyframe loses you a chunk of video. Therefore, we would like it if our router kept keyframes in preference to non-keyframes.

Furthermore, there’s been research on Quality of Service, by which we might somehow reserve a certain level of network performance. QoS is an active research area, but one thing is for certain, to achieve it we definitely need to discriminate between packets.

Network neutrality advocates are concerned that the people who own the routers might choose to discriminate between packets on some basis that’s bad, like “Google didn’t pay me any money, so I’ll deliver their packets slowly or not at all.” The thing is, exactly no one is proposing to do this. Were any ISP to do it (and it is the ISPs that people seem particularly concerned about), they would immediately lose their customers to another ISP that did not.

But what an ISP might do is establish some preferences between different sorts of traffic; for example to prefer interactive traffic over large downloads. (According to the article, Comcast has done this.) Someday, they might even implement Quality of Service. This is all for the good. It would be a tragedy if network neutrality were to prevent it.

Google argues for neutrality this way:

The broadband carriers should not be permitted to use their market power to discriminate against competing applications or content.

If I had ever heard of such a thing happening, I might feel differently. Until then, network neutrality is a solution to a non-existent problem, and one that, if not implemented very carefully, could be very harmful to the development of the Internet. Readers of this blog will not be surprised that I would estimate the likelihood of Congress being so careful at roughly zero.

DARPA has awarded contracts to several firms to build prototypes for the Vulture, an unmanned plane that would stay aloft for five years on solar power:

What the Pentagon wants is essentially a maneuverable satellite replacement: a fixed-wing, heavier-than-air craft that’s high enough to “see” large swaths of the Earth at once, but one that can also reposition itself to circle over new areas of interest, something satellites in fixed orbits can’t do.

It has to be able carry a 1,000-pound payload, battle stratospheric winds, generate a continuous 5 kilowatts of electricity — and it can’t use nuclear power to do so.

“We want to completely change the paradigm of how we think of aircraft,” Vulture project manager Daniel Newman tells Flight. “We would no longer define an aircraft by the launch, recover, maintain, launch cycle.”

Cool.  It also would fly at 90 thousand feet, higher than most planes but quite a bit lower than satellites, which would give it better resolution.

The National Post has an column by Lawrence Solomon on another edit war at Wikipedia. (Via Instapundit.) In this case, the page is about history professor Naomi Oreskes’s notorious essay in which she reports that she did a keyword search on “global climate change” and found no papers that disagreed with a human origin for climate change. The edit war regarded another analysis in which Benny Peiser, a social anthropologist, attempted to reproduce Oreskes’s results but obtained strikingly different results instead. Solomon writes that the article implied that Peiser has retracted his critique (which is manifestly not the case), but Solomon’s efforts to correct the article were repeatedly reverted.

My scan of the edits reveals that the edit history is a bit more complicated than Solomon’s column lets on. In the end, however, the entire discussion of Peiser’s critique was deleted, thus removing the misleading information regarding Peiser himself, but leaving the critique unexpressed. This was done while the page was supposedly protected.

While investigating, I learned a few interesting things. I’ve always known that Wikipedia can’t really be trusted on matters of controversy (e.g., climate change, NFL quarterbacks, or even the moon landings), but I had thought that one could get some idea of the arguments by reading the talk pages. It turns out, however, that people “referee” the talk pages as well, even there deleting comments they don’t like.

I also learned that the major Wikipedia editors don’t like to follow the rules that apply to everyone else. The antagonist in Solomon’s story violated the three-revert rule and was not apparently held to account. In fact, they even have a policy that says it’s rude to remind regulars to follow the rules.

Any collaborative project needs to deal with a diversity of interests. In a typical open source project, there are two main interests, the developers and the vandals. It’s not hard to see that one interest is legitimate and the other is not. It’s not so for Wikipedia. As a major source of information (particularly since Google gives it a special status), it has many non-vandal interests, and those interests can’t agree on which are legitimate. Plus there’s the iron law of bureaucracy. This makes Wikipedia into a competition, rather than a collaboration, which explains it in a nutshell.

The estimate of the Bakken oil deposit is out, and it’s smaller than some had hoped. It’s still a significant amount, but by no means OPEC-crushing. (Via Instapundit.)

(Previous post.)

Business Week reports:

A long-awaited federal report on oil that could be recovered in parts of North Dakota, Montana and two Canadian provinces is to be released this week. . .

In 1995, the Geological Survey estimated that using technology available at that time, 151 million barrels of oil could be recovered in the Bakken, said Brenda Pierce, a geologist and program coordinator for the agency’s energy resources program.

Pierce said she would not disclose the study’s findings until Thursday. Asked whether the estimate would be an increase from the 1995 figure, she said, “There is industry in there and having success. There’s your answer.”

(Via Instapundit.)

Looks like this is for real.

(Previous post.)

UPDATE (5/20): For some reason, a lot of people are finding this post through search engines. Here’s my latest post on the topic.

The London Times has an atrocious article about how “the grid” may soon make the Internet obsolete:

The Internet could soon be made obsolete. The scientists who pioneered it have now built a lightning-fast replacement capable of downloading entire feature films within seconds. At speeds about 10,000 times faster than a typical broadband connection, “the grid” will be able to send the entire Rolling Stones back catalogue from Britain to Japan in less than two seconds.

They seem to be falling into the trap of thinking of the Internet as a bunch of wires.  In fact, the Internet is basically just an algorithm for routing packets, so a replacement would have to be a new, better way to route packets. That’s not apparently what they’re talking about here. (I say “apparently,” because it’s not at all clear what they actually are talking about.)

The idea of grid computing (decentralized networked computing ventures, often consisting of volunteers) seems like a good way to handle the amount of data they expect to generate at the Large Hadron Collider, but it’s hardly new.

So what is new?  If we assume that there’s anything to do this article at all, the CERN guys are facing the problem of getting their data out to their grid participants.  (In most grid applications, such as SETI@Home, participants conduct large computations on small amounts of data, so this isn’t an issue.)  Here, a colleague tells me, they’re talking about a content delivery network using some dedicated bandwidth.

Sounds like a fine approach, but not a replacement for the Internet.  (In fact, they probably use IP to route packets over their network, which would make it actually part of the Internet.)  Semantics aside, it also doesn’t sound like something that will have any impact on most people lives, since the people who paid for the dedicated bandwidth are unlikely to let people use it to download films.

(ASIDE: What would be great would be if someone were to come up with a way for ordinary people to exploit the grid computing paradigm. . .)

Why on earth can’t the iPhone synchronize bookmarks with Firefox?

The YAL-1 Airborne Laser, explained in a Boeing promotional video.  (Warning: fairly cheesy.)  Apparently they’ve licked the problem of atmospheric distortion by analyzing the return from a tracking laser before firing the high-energy laser.

Another use for the iPhone: Sometimes it’s just too much trouble to get up from your armchair and walk two steps to your computer.

Ever wanted to be a fly on the wall somewhere?  Soon you can.

The system requirements for Crysis:

OS - Windows XP or Windows Vista

Processor - 2.8 GHz or faster (XP); 3.2 GHz or faster (Vista)

Memory - 1.0 GB RAM or 1.5 GB RAM (Vista)

The New Republic has an article on edit wars at Wikipedia on the pages for Clinton and Obama. (Via The New Editor, via Instapundit.) No surprise: both sides have some juveniles. But the key point is here, I think:

The battles over Hillary’s and Obama’s pages have been so heated because the stakes are so high. The candidates’ Wikipedia pages are their second Google hits, right after their official campaign portals.

Wikipedia almost always comes near the top on Google. Why? A friend at Google once confirmed for me what I suspected, that it’s not simply the result of their page ranking algorithm — they put it there deliberately. “People like Wikipedia,” he said.

True enough, people like Wikipedia, but it tends to be of limited use for controversial subjects. Moreover, you can be surprised by what turns out to be controversial; I once read their article about the “alleged moon landings.” (This has long since been fixed.) I’d rather they simply applied their algorithm and let the chips fall where they may.

I first saw this piece on a huge oil discovery in North Dakota a little while ago, but didn’t pay much attention, since I’d never heard of Next Energy News before. Now that piece is making the rounds of the blogosphere (Rand Simberg, Instapundit) and one of Simberg’s commenters points to this post, which strikes me as credible. (As the very least, it has lots of supporting links.)

If this pans out, the upshot is 100–300 billion barrels of oil in the Williston Basin, which spans the U.S.-Canada border (with the largest portions being in North Dakota and Saskatchewan). By comparison, Saudi Arabia has 260 billion barrels in proven oil reserves.

UPDATE (5/20): For some reason, a lot of people are finding this post through search engines. Here’s my latest post on the topic.

This morning my post on Blackwater Fever was Instalanched.  My first response was to use it to try to draw attention to my ad for John McCain.  My second response was to wonder how on earth Glenn Reynolds found it.  Before today, my blog was known to fewer than 10 people, and I’m not actively promoting it.  According to the stats, I’ve received no traffic from search engines.  So he must have found out via a linkback.

The thing is, Reynolds must get thousands of linkbacks every day.  He can’t possibly look at them all, and also scour the Internet for interesting articles, and also hold down a job as a law professor, can he?  I’m curious, so I’m going to try an experiment:  I have another link to Instapundit in this post.  If I get another Instalanche, we will have empirically confirmed that Glenn Reynolds has no life.

UPDATE: Reynolds refuses to be manipulated into a sending a second Instalanche, but he (or someone at UT claiming to be him) weighs in in the comments.  Hypothesis confirmed!

Under pressure from Islamists, Network Solutions shuts down a website being used to promote an upcoming anti-Islamic movie.  (Via Instapundit.)  Apparently Network Solutions has a policy against “objectionable material of any kind or nature.”

Hmm, perhaps I can get them to shut down websites promoting aspect-oriented programming.  Get your censorship while the getting’s good.